see this data set + chp on diaxic shift in gcr1 mutants
paper on lysine pathway
Setup environment
Display design and replicate tally
cluster significant genes by their profiles
#> [1] "Intercept"
#> [2] "genotype_EDS1_vs_WT"
#> [3] "genotype_EDS1_LYS14_vs_WT"
#> [4] "genotype_EDS1_RGT1_vs_WT"
#> [5] "genotype_LYS14_vs_WT"
#> [6] "genotype_RGT1_vs_WT"
#> [7] "aminoAcid_HisMetLeuUra_vs_LysHisMetLeuUra"
#> [8] "genotypeEDS1.aminoAcidHisMetLeuUra"
#> [9] "genotypeEDS1_LYS14.aminoAcidHisMetLeuUra"
#> [10] "genotypeEDS1_RGT1.aminoAcidHisMetLeuUra"
#> [11] "genotypeLYS14.aminoAcidHisMetLeuUra"
#> [12] "genotypeRGT1.aminoAcidHisMetLeuUra"#> Saving 7 x 5 in image
Plot gene with the most significant effect btwn aminoAcid groups
ptchwrk = plt1 + plt2+ plt4 + plot_layout(guides = 'collect')
gene_name = "THI4"
gene_id = 'YGR144W'
ptchwrk = ptchwrk + plot_annotation(
title = gene_name,
subtitle = gene_id
) & theme_minimal()
# ggsave(here("plots/condition_specific_THI4.png"), ptchwrk, device = "pdf")
ptchwrk
# re-do this with new venn list
# intersect(gene_lists_for_venn$down$singles$RGT1, gene_lists_for_venn$up$singles$EDS1)Genes which change in different directions
There are no genes which change in different directions in EDS1 and RGT1.
YPL187W is up in the EDS1 KO and down in LYS14.
Mating pheromone alpha-factor, made by alpha cells; interacts with mating type a cells to induce cell cycle arrest and other responses leading to mating; also encoded by MF(ALPHA)2, although MF(ALPHA)1 produces most alpha-factor; binds copper(II) ions 3 4 5 6 8
- YBR296C, PHO89 is down in EDS1 and up in LYS14
Plasma membrane Na+/Pi cotransporter; active in early growth phase; similar to phosphate transporters of Neurospora crassa; transcription regulated by inorganic phosphate concentrations and Pho4p; mutations in related human transporter genes hPit1 and hPit2 are associated with hyperphosphatemia-induced calcification of vascular tissue and familial idiopathic basal ganglia calcification 1 2 3 4
- there are no genes which change in different directions between RGT1 and LYS14
Lysine Pathway
Lysine Biosynthesis Genes–Revist Eds1 activates LYS4, ACO2, LYS9, LYS12, CTP1, LYS2, LYS1, LYS14 in low glucose -lysine. LYS4 repressed in galactose +/-lysine, ACO2 repressed in galactose -lysine, LYS2 repressed in high glucose. LYS4, LYS9, ACO2 are the only genes for which |LFC| > 1.
getSetHeatMap(long_res_df, gene_set_df_list$lysine_genes,
c("EDS1_", "RGT1_", "EDS1_RGT1_", "LYS14_", "EDS1_LYS14_"),
c('plus_lys', 'minus_lys'), "lysine genes -- unfiltered for lfc or padj")#> Joining, by = "id"
same with LYS
Eds1 has a repressive efect on glyoxylate cycle genes – combine with TCA network
Eds1 has a repressive effect on TCA cycle genes
TCA genes should be ‘on’ in WT galactose. If the TCA genes go up when EDS1 is KO’d, then EDS1 does have a repressive effect on TCA genes. Another factor to this, however, is that overtime in galactose, we probably expect TCA genes to go up.
getSetHeatMap(long_res_df, gene_set_df_list$tca_genes,
c("EDS1_", "RGT1_", "EDS1_RGT1_", "LYS14_"),
c('plus_lys', 'minus_lys'), "tca cycle genes -- unfiltered for lfc or padj")#> Joining, by = "id"
getSetHeatMap(long_res_df, gene_set_df_list$membrane_transport_genes,
c("EDS1_", "RGT1_", "EDS1_RGT1_", "LYS14_"),
c('plus_lys', 'minus_lys'),
"membrane transport -- unfiltered for lfc or padj")#> Joining, by = "id"
getSetHeatMap(long_res_df, gene_set_df_list$response_to_glucose_genes,
c("EDS1_", "RGT1_", "EDS1_RGT1_", "LYS14_"),
c('plus_lys', 'minus_lys'),
"response to glucose -- unfiltered for lfc or padj")#> Joining, by = "id"
Eds1 directly represses SKS1 in both galactose +/- lysine.
SKS1 is a repressor kinase of SNF3 (Yeast Genome), a low glucose sensor required for induction of certain hexose transporters and which triggers inhibition of Rgt1 (Ozcan et al, 1998).
CHASE NOTE: I want to review that SKS1 is a repressor of SNF3. SGD says it is indepdendent of the SNF3 pathway, and that SKS1 is required in low gluc in SNF3 null strains. Confusing b/c SKS means suppressor kinase of snf3.
The snf3 genomic suppressors, rgt1, RGT2 and ssn6, are also capable of suppressing snf3 associated growth defects in a strain lacking sks1.
getSetHeatMap(long_res_df, gene_set_df_list$hexose_transmem_trans_genes,
c("EDS1_", "RGT1_", "EDS1_RGT1_", "LYS14_"),
c('plus_lys', 'minus_lys'),
"hexose tarnsmembrane transporter genes -- unfiltered for lfc or padj")#> Joining, by = "id"
getSetHeatMap(long_res_df, gene_set_df_list$glucose_transmem_trans_genes,
c("EDS1_", "RGT1_", "EDS1_RGT1_", "LYS14_"),
c('plus_lys', 'minus_lys'),
"glucose transmembrane transport genes -- unfiltered for lfc or padj")#> Joining, by = "id"
getSetHeatMap(long_res_df, gene_set_df_list$glucose_transmem_transport_activity_genes,
c("EDS1_", "RGT1_", "EDS1_RGT1_", "LYS14_"),
c('plus_lys', 'minus_lys'),
"glucose transmem trans activity -- unfiltered for lfc or padj")#> Joining, by = "id"
getSetHeatMap(long_res_df, gene_set_df_list$glucose_import_genes,
c("EDS1_", "RGT1_", "EDS1_RGT1_", "LYS14_"),
c('plus_lys', 'minus_lys'),
"glucose import -- unfiltered for lfc or padj")#> Joining, by = "id"
getSetHeatMap(long_res_df, gene_set_df_list$hexokinase_genes,
c("EDS1_", "RGT1_", "EDS1_RGT1_", "LYS14_"),
c('plus_lys', 'minus_lys'),
"hexokinase -- unfiltered for lfc or padj")#> Joining, by = "id"
getSetHeatMap(long_res_df, gene_set_df_list$glyoxylate_genes,
c("EDS1_", "RGT1_", "EDS1_RGT1_", "LYS14_"),
c('plus_lys', 'minus_lys'), "glyoxylate genes -- unfiltered for lfc or padj")#> Joining, by = "id"
Fermentation pathways?
These should be ‘off’ in galactose. If perturbing the genes in question cause fermentation pathway genes to increase, then it is evidence that these genes have a role in fermentation
getSetHeatMap(long_res_df, gene_set_df_list$anaerobic_resp_genes,
c("EDS1_", "RGT1_", "EDS1_RGT1_", "LYS14_"),
c('plus_lys', 'minus_lys'), "anaerobic respiration -- unfiltered for lfc or padj")#> Joining, by = "id"
Aerobic respiration pathways
These should be ‘on’ in galactose, presumably. Down regulation would point to genes that a given perturbed locus has a positive impact on. up regulation points to genes which a given perturbed locus is supressing. Expecting few of those?
getSetHeatMap(long_res_df, gene_set_df_list$aerobic_resp_genes,
c("EDS1_", "RGT1_", "EDS1_RGT1_", "LYS14_"),
c('plus_lys', 'minus_lys'), "aerobic respiration -- unfiltered for lfc or padj")#> Joining, by = "id"
getSetHeatMap(long_res_df, gene_set_df_list$glycolytic_genes,
c("EDS1_", "RGT1_", "EDS1_RGT1_", "LYS14_"),
c('plus_lys', 'minus_lys'), "glycolysis genes -- unfiltered for lfc or padj")#> Joining, by = "id"
Eds1 strongly and directly activates HXK1, where Rgt1 does not.
Eds1 directly activates CDC19 in galactose +lysine.
CDC19 synthesizes pyruvate from phosphoenolpyruvate, the last step before TCA. Rgt1 directly represses CDC19 in our -lysine conditions, but activates in galactose +lysine.
CHASE QUESTION: why would the presence/absence of lysine affect TCA cycle?
Eds1 and Rgt1 directly regulate many common targets in the same direction
venn$plots$minus_lys$up$singles
venn$plots$minus_lys$down$singles
venn$plots$plus_lys$up$singles
venn$plots$plus_lys$down$singles
intersect top up/down genes in eds1/rgt1, do GO
Do the same with complements
Reguation of Eds1 by RGT1 or LYS14?
bind_rows(shrunken_res_lists$plus_lys$LYS14 %>%
as_tibble(rownames = 'gene') %>%
select(gene, log2FoldChange, padj) %>%
mutate(condition = 'plus_lys',
genotype = 'LYS14'),
shrunken_res_lists$minus_lys$LYS14 %>%
as_tibble(rownames = 'gene') %>%
select(gene, log2FoldChange, padj) %>%
mutate(condition = 'minus_lys',
genotype = 'LYS14')
) %>%
filter(gene %in% c('YKL038W', 'YBR033W')) %>%
ggplot() +
geom_bar(aes(interaction(condition, genotype), log2FoldChange, fill = gene), stat='identity', position = 'dodge') + ylim(-.5, .5)
dds_vst = varianceStabilizingTransformation(dds, blind=FALSE)
assays(dds_vst)[[1]] %>%
as_tibble(rownames = 'gene') %>%
pivot_longer(-gene, names_to = 'sample', values_to = 'vst') %>%
left_join(as_tibble(colData(dds_vst))) %>%
filter(gene %in% c('YKL038W', 'YBR033W'), genotype %in% c('WT', "LYS14")) %>%
ggplot() +
geom_boxplot(aes(interaction(genotype, aminoAcid), vst, fill = gene)) +
geom_point(aes(interaction(genotype, aminoAcid), vst))#> Joining, by = "sample"
HXT expression
heatmap of all HXT genes
RGT1/MIG pathway genes
heatmap of these genes
Eds1 directly represses MTH1 in galactose with lysine. This needs binding data
Rgt1 represses in galactose with lysine, galactose without lysine, high glucose late.
Mig1 represses in high glucose early, high glucose late. Mig2 activates in high glucose early, represses in high glucose late. Rgt1, Mig1, Mig2 all represses MTH1 according to Kuttykrishnan et al. Eds1 indirectly represses STD1 in galactose with lysine and galactose without lysine. Rgt1 represses STD1 in galactose with lysine, galactose without lysine, and low glucose. Eds1 directly represses MIG1 in galactose no lysine, activates in glucose. Rgt1 represses MIG1 in galactose +/- lysine.
plotGeneCountCustom(dds, 'YBR033W', 'EDS1')$combined#> Warning: Removed 3 rows containing non-finite values (stat_summary).#> Warning: Removed 3 rows containing missing values (geom_point).#> Warning: Removed 5 rows containing non-finite values (stat_summary).#> Warning: Removed 5 rows containing missing values (geom_point).#> Warning: Removed 4 rows containing non-finite values (stat_summary).#> Warning: Removed 4 rows containing missing values (geom_point).
plotGeneCountCustom(dds, 'YKL038W', 'RGT1')$combined#> Warning: Removed 5 rows containing non-finite values (stat_summary).#> Warning: Removed 5 rows containing missing values (geom_point).#> Warning: Removed 4 rows containing non-finite values (stat_summary).#> Warning: Removed 4 rows containing missing values (geom_point).
plotGeneCountCustom(dds, 'YGL035C', 'MIG1')$combined
plotGeneCountCustom(dds, 'YGL209W', 'MIG2')$combined
plotGeneCountCustom(dds, 'YGL209W', 'MIG2')$combined
Phosphate Regulation
Eds1 causes highly significant repression of PHO11, PHO12, PHO89 (11-fold), SPL2 (32-fold) across all time points.
Note: not so true in galac, though there is a down reg and what seems to be an additive effect
no evidence of direct binding
Rgt1 also causes highly significant repression of these genes in glucose -lysine. remains to be checked
plotGeneCountCustom(dds, 'YBR296C', 'PHO89')$combined
plotGeneCountCustom(dds, 'YAR071W', 'PHO11')$combined
plotGeneCountCustom(dds, 'YHR215W', 'PHO12')$combined
plotGeneCountCustom(dds, 'YHR136C', 'SPL2')$combined
getSetHeatMap(long_res_df, gene_set_df_list$acid_phosphate_activity,
c("EDS1_", "RGT1_", "EDS1_RGT1_", "LYS14_"),
c('plus_lys', 'minus_lys'), "acid phosphate activity -- unfiltered for lfc or padj")#> Joining, by = "id"
getSetHeatMap(long_res_df, gene_set_df_list$phosphate_ion_transmem_transport,
c("EDS1_", "RGT1_", "EDS1_RGT1_", "LYS14_"),
c('plus_lys', 'minus_lys'), "phosphate_ion_transmem_transport -- unfiltered for lfc or padj")#> Joining, by = "id"
getSetHeatMap(long_res_df, gene_set_df_list$phoscontaining_compound_meta_proc,
c("EDS1_", "RGT1_", "EDS1_RGT1_", "LYS14_"),
c('plus_lys', 'minus_lys'), "phoscontaining_compound_meta_proc -- unfiltered for lfc or padj")#> Joining, by = "id"
Transcription factors
Eds1 directly activates Mig3 in high glucose -lysine.
plotGeneCountCustom(dds, 'YER028C', 'MIG3')$combined
Eds1 directly represses MIG1 in galactose no lysine, activates in glucose. (see plot above)
Rgt1 represses MIG1 in galactose +/- lysine. (see plot above)
Eds1 directly represses MTH1 in galactose with lysine.
plotGeneCountCustom(dds, 'YDR277C', 'MTH1')$combined
NOTES
diaxic shift
Following glucose exhaustion, cells switch to oxygenic respiration (a.k.a. “diauxic shift”), which allows catabolizing ethanol and the other carbon compounds via the TCA cycle and oxidative phosphorylation in the mitochondria.
MIG1 represses genes whose expression is shut off when glucose is present such as those enconding enzymes for sugars maltose, sucrose and galactose
When cells become limited for glucose, Mig1p is phosphorylated by the Snf1 kinase complex, composed of the Snf1 kinase catalytic subunit, the gamma subunit Snf4p, and a beta subunit encoded by SIP1, SIP2, or GAL83 (5).